Lawlor — Primate Ussher\s Library before 16Ij.l. 257 



shown that a volume is common to A and C,^ and in eight that a 

 volume is common to A and B,'^ while six volumes are registered in all 

 three catalogues.^ All these books are now in Trinity College. So 

 also are fifteen voliunes named in A which are not proved by the 

 press-marks to have been in B or C* To the conclusion which seems 

 to follow of necessity fi'om these premises I conceive that only one 

 objection is possible. Assuming the correctness of our inference, and 

 that the proper chronological order of the catalogues is A, B, C,^ we 

 should expect that a volume included in A, and still preserved in the 

 Trinity College Libraiy, would be mentioned in B and C. And in 

 like manner, a volume in B ought also to have a place in C. But in 

 fact we find that, in many instances, oui' table does not prove that this 

 is so. To take a concrete case, No. 26 was certainly in A and C, why 

 then if B is a catalogue of the same collection is its press-mark absent 

 fi'om our second column ? In other words, we have to account for the 

 blank spaces in our second and third columns. This can be done 

 without diffictdty. In the third column there are only nine blanks of 

 the kind mentioned. We may suppose that the writer of C accidently 

 omitted to register the volumes, or that he described them in such a 

 way that I have not been able to identify them, or finally, that they 

 had been borrowed after catalogue B was prepared, and were not 

 returned till after C had been drawn up. In the second column the 

 blanks are much more nujnerous, but this was to be expected. B is a 

 local, not an alphabetical, catalogue. It is, therefore, impossible to 

 find a volume in it unless its press-mark is known beforehand. And, 

 the press-marks can only be ascertained, apart from the catalogues, if 

 they are preserved in the volumes recorded in our fifth column. I^ow, 

 the majority of these vohxmes have been re-bound, and in the process 

 of binding, old marks are in great danger of being destroyed. Actually 

 in the case of the press-marks given in B, this danger is much greater 

 than in the case of those given in C, although many of the latter have 

 been lost. In the first place, the B marks are older than the C marks, 

 and between the two, as we shall see, there came an interval during 



1 JSTos. 7, 8, 9, 10, 26, 31, 35, 36, 38, 39, 41, 47, 62, 63, 65. 



2 Nos. 10, 36, 38, 39, 47, 59, 60, 65. 



3 Nos. 10, 36, 38, 39, 47, 65. 



* Nos. 1, 6, 13, 20, 28, 30, 32, 50, 52, 54, 67, 58, 68, 71, 73. 



^ Proof will be given of the truth of this assumption lower down. But, of 

 course, a similar objection may be raised on any supposition as to the order of the 

 catalogues in time, provided either B or C is later than A. 



